Struct SampledStatistics 

pub struct SampledStatistics { /* private fields */ }

Approximate statistics computed from a random sample of the triple store.

Useful for large graphs where computing exact statistics is too expensive. Uses reservoir sampling semantics — the caller feeds triples and the structure maintains a representative sample of size sample_size.

Implementations

impl SampledStatistics

pub fn new(sample_size: usize) -> Self

Create a new sampler that will keep at most sample_size triples.

pub fn observe(&mut self, subject: &str, predicate: &str, object: &str)

Feed a triple to the sampler. Uses reservoir sampling to decide whether to keep it.

Because true reservoir sampling requires a PRNG, this implementation uses a deterministic hash-based pseudo-random selection so that tests are reproducible without an external RNG.

pub fn total_seen(&self) -> u64

Number of triples seen so far (population size).

pub fn sample_size(&self) -> usize

Number of triples in the sample.

pub fn sample(&self) -> &[(String, String, String)]

Access the raw sample.

pub fn estimated_total(&self) -> u64

Estimated total triples (same as total_seen, since each observe is one triple).

pub fn estimated_distinct_predicates(&self) -> u64

Estimated number of distinct predicates, extrapolated from the sample.

pub fn estimate_cardinality( &self, subject: Option<&str>, predicate: Option<&str>, object: Option<&str>, ) -> u64

Estimated cardinality for a triple pattern using sample-based selectivity.

pub fn graph_stats(&self) -> &GraphStatistics

Access the underlying sample-based GraphStatistics.

Trait Implementations

impl Clone for SampledStatistics

fn clone(&self) -> SampledStatistics

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for SampledStatistics

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.